Re: [PATCH 1/3] perf/core: Flush PMU internal buffers for per-CPU events

2020-11-25 Thread Michael Ellerman
Namhyung Kim writes: > Hello, > > On Mon, Nov 23, 2020 at 8:00 PM Michael Ellerman wrote: >> >> Namhyung Kim writes: >> > Hi Peter and Kan, >> > >> > (Adding PPC folks) >> > >> > On Tue, Nov 17, 2020 at 2:01 PM Namhyung Kim wrote: >> >> >> >> Hello, >> >> >> >> On Thu, Nov 12, 2020 at 4:54 AM

Re: [PATCH 1/3] perf/core: Flush PMU internal buffers for per-CPU events

2020-11-24 Thread Liang, Kan
On 11/24/2020 12:42 AM, Madhavan Srinivasan wrote: On 11/24/20 10:21 AM, Namhyung Kim wrote: Hello, On Mon, Nov 23, 2020 at 8:00 PM Michael Ellerman wrote: Namhyung Kim writes: Hi Peter and Kan, (Adding PPC folks) On Tue, Nov 17, 2020 at 2:01 PM Namhyung Kim wrote: Hello, On

Re: [PATCH 1/3] perf/core: Flush PMU internal buffers for per-CPU events

2020-11-23 Thread Madhavan Srinivasan
On 11/24/20 10:21 AM, Namhyung Kim wrote: Hello, On Mon, Nov 23, 2020 at 8:00 PM Michael Ellerman wrote: Namhyung Kim writes: Hi Peter and Kan, (Adding PPC folks) On Tue, Nov 17, 2020 at 2:01 PM Namhyung Kim wrote: Hello, On Thu, Nov 12, 2020 at 4:54 AM Liang, Kan wrote: On

Re: [PATCH 1/3] perf/core: Flush PMU internal buffers for per-CPU events

2020-11-23 Thread Namhyung Kim
Hello, On Mon, Nov 23, 2020 at 8:00 PM Michael Ellerman wrote: > > Namhyung Kim writes: > > Hi Peter and Kan, > > > > (Adding PPC folks) > > > > On Tue, Nov 17, 2020 at 2:01 PM Namhyung Kim wrote: > >> > >> Hello, > >> > >> On Thu, Nov 12, 2020 at 4:54 AM Liang, Kan > >> wrote: > >> > > >> >

Re: [PATCH 1/3] perf/core: Flush PMU internal buffers for per-CPU events

2020-11-23 Thread Michael Ellerman
Namhyung Kim writes: > Hi Peter and Kan, > > (Adding PPC folks) > > On Tue, Nov 17, 2020 at 2:01 PM Namhyung Kim wrote: >> >> Hello, >> >> On Thu, Nov 12, 2020 at 4:54 AM Liang, Kan wrote: >> > >> > >> > >> > On 11/11/2020 11:25 AM, Peter Zijlstra wrote: >> > > On Mon, Nov 09, 2020 at

Re: [PATCH 1/3] perf/core: Flush PMU internal buffers for per-CPU events

2020-11-20 Thread Namhyung Kim
Hi Peter and Kan, (Adding PPC folks) On Tue, Nov 17, 2020 at 2:01 PM Namhyung Kim wrote: > > Hello, > > On Thu, Nov 12, 2020 at 4:54 AM Liang, Kan wrote: > > > > > > > > On 11/11/2020 11:25 AM, Peter Zijlstra wrote: > > > On Mon, Nov 09, 2020 at 09:49:31AM -0500, Liang, Kan wrote: > > > > > >>

Re: [PATCH 1/3] perf/core: Flush PMU internal buffers for per-CPU events

2020-11-16 Thread Namhyung Kim
Hello, On Thu, Nov 12, 2020 at 4:54 AM Liang, Kan wrote: > > > > On 11/11/2020 11:25 AM, Peter Zijlstra wrote: > > On Mon, Nov 09, 2020 at 09:49:31AM -0500, Liang, Kan wrote: > > > >> - When the large PEBS was introduced (9c964efa4330), the sched_task() > >> should > >> be invoked to flush the

Re: [PATCH 1/3] perf/core: Flush PMU internal buffers for per-CPU events

2020-11-11 Thread Liang, Kan
On 11/11/2020 11:25 AM, Peter Zijlstra wrote: On Mon, Nov 09, 2020 at 09:49:31AM -0500, Liang, Kan wrote: - When the large PEBS was introduced (9c964efa4330), the sched_task() should be invoked to flush the PEBS buffer in each context switch. However, The perf_sched_events in

Re: [PATCH 1/3] perf/core: Flush PMU internal buffers for per-CPU events

2020-11-11 Thread Peter Zijlstra
On Mon, Nov 09, 2020 at 09:49:31AM -0500, Liang, Kan wrote: > - When the large PEBS was introduced (9c964efa4330), the sched_task() should > be invoked to flush the PEBS buffer in each context switch. However, The > perf_sched_events in account_event() is not updated accordingly. The >

Re: [PATCH 1/3] perf/core: Flush PMU internal buffers for per-CPU events

2020-11-09 Thread Liang, Kan
On 11/9/2020 12:33 PM, Peter Zijlstra wrote: On Mon, Nov 09, 2020 at 09:49:31AM -0500, Liang, Kan wrote: Maybe we can frob x86_pmu_enable()... Could you please elaborate? Something horrible like this. It will detect the first time we enable the PMU on a new task (IOW we did a context

Re: [PATCH 1/3] perf/core: Flush PMU internal buffers for per-CPU events

2020-11-09 Thread Peter Zijlstra
On Mon, Nov 09, 2020 at 09:49:31AM -0500, Liang, Kan wrote: > - The patch 1 tries to fix broken per-CPU events. The CPU context cannot be > retrieved from the task->perf_event_ctxp. So it has to be tracked in the > sched_cb_list. Yes, the code is very similar to the original codes, but it > is

Re: [PATCH 1/3] perf/core: Flush PMU internal buffers for per-CPU events

2020-11-09 Thread Peter Zijlstra
On Mon, Nov 09, 2020 at 09:49:31AM -0500, Liang, Kan wrote: > > Maybe we can frob x86_pmu_enable()... > > Could you please elaborate? Something horrible like this. It will detect the first time we enable the PMU on a new task (IOW we did a context switch) and wipe the counters when user RDPMC is

Re: [PATCH 1/3] perf/core: Flush PMU internal buffers for per-CPU events

2020-11-09 Thread Liang, Kan
On 11/9/2020 6:04 AM, Peter Zijlstra wrote: On Mon, Nov 09, 2020 at 10:52:35AM +0100, Peter Zijlstra wrote: On Fri, Nov 06, 2020 at 01:29:33PM -0800, kan.li...@linux.intel.com wrote: From: Kan Liang Sometimes the PMU internal buffers have to be flushed for per-CPU events during a context

Re: [PATCH 1/3] perf/core: Flush PMU internal buffers for per-CPU events

2020-11-09 Thread Peter Zijlstra
On Mon, Nov 09, 2020 at 10:52:35AM +0100, Peter Zijlstra wrote: > On Fri, Nov 06, 2020 at 01:29:33PM -0800, kan.li...@linux.intel.com wrote: > > From: Kan Liang > > > > Sometimes the PMU internal buffers have to be flushed for per-CPU events > > during a context switch, e.g., large PEBS.

Re: [PATCH 1/3] perf/core: Flush PMU internal buffers for per-CPU events

2020-11-09 Thread Peter Zijlstra
On Fri, Nov 06, 2020 at 01:29:33PM -0800, kan.li...@linux.intel.com wrote: > From: Kan Liang > > Sometimes the PMU internal buffers have to be flushed for per-CPU events > during a context switch, e.g., large PEBS. Otherwise, the perf tool may > report samples in locations that do not belong to

[PATCH 1/3] perf/core: Flush PMU internal buffers for per-CPU events

2020-11-06 Thread kan . liang
From: Kan Liang Sometimes the PMU internal buffers have to be flushed for per-CPU events during a context switch, e.g., large PEBS. Otherwise, the perf tool may report samples in locations that do not belong to the process where the samples are processed in, because PEBS does not tag samples